2016 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
such as the capability to process particles as small as 15–20
microns, sometimes even finer. Their simplicity, robustness,
high capacity, and minimal operator attention further con-
tribute to their appeal (Falconer, 2003). These competitive
advantages have prompted the utilization of Falcon concen-
trators across various materials and objectives (Farajzadeh
&Chehreh Chelgani, 2022). Examples include gold recov-
ery (Lins et al., 1992), coal cleaning (Honaker et al., 1996),
pre-enrichment and desliming for tungsten concentration
from skarns (Foucaud et al., 2019), recycling (Duan et al.,
2009 Zhang et al., 2018), and mine tailings reprocessing
(Filippov et al., 2016), as well as cassiterite recovery from
pegmatites (Deveau, 2016).
This article explores the application of the Falcon
concentrator to lithium ore from the Beauvoir granite, a
well-documented lithium (Li) rare metal deposit in France
known for its noteworthy tin (Sn), niobium (Nb), and
tantalum (Ta) concentrations France (Cuney et al., 1992
EMILI, 2023 Gourcerol et al., 2019 Linnen et al., 2012).
The primary objective is to evaluate the Falcon concentra-
tor’s potential in recovering Sn, Nb, and Ta as by-products
of Li. The study introduces recovery and enrichment ratio
models to identify optimal operating conditions. The find-
ings are then discussed with the aim of providing new
insights into the role of machine parameters on the separa-
tion mechanisms in this kind of separator.
MATERIAL AND METHODS
Materials
The ore utilized in this investigation was provided by
Imerys Ceramics France and consist in fresh granite from
the Beauvoir rare metal granite. Obtained during an explo-
ration campaign in 2018 from the central region of the
deposit, the core samples have been extensively character-
ized and labeled as PERC C by Demeusy and colleagues
(Demeusy et al., 2023). The composite sample obtained
is composed at 95% albite, quartz, micas, and minor
K-feldspar, with additional phases constituting the remain-
ing 5%, including phosphates, fluorite, beryl, kaolinite,
cassiterite, and columbo-tantalite (Cuney et al., 1992). The
measured metal concentrations within the sample are as fol-
lows: approximately 0.89% Li2O, 1073 ppm Sn, 155 ppm
Ta, and 106 ppm Nb (Demeusy et al., 2023). Liberation
size of the Li bearing lepidolite occurs at 300 µm and the
cassiterite/Colombo-tantalite at 100 µm.
Sample Preparation
The ore processing involved a series of crushing and grind-
ing steps. Initially, half cores were successively crushed using
a primary jaw crusher, followed by a secondary jaw crusher,
and ultimately a tertiary gyratory crusher, achieving a final
particle size of minus 2mm. Subsequently, the crushed
product underwent classification and further grinding to a
top size of 280µm. This grinding process utilized a closed-
loop system with a 60 cm diameter sieve, coupled with a
40 cm diameter open circuit tubular ball mill operating
at 70% of its critical speed and a solid content of 65%.
Following the grinding process, the resulting product was
subjected to drying and then sampled and split in 1 kg bags
for subsequent gravity separation trials.
Feed Characteristics
The minus280µm process feed exhibited a size distribu-
tion with a P50 (median particle size) of 100 µm and a P80
(particle size at 80% passing) of 200 µm (Figure 1). Upon
conducting back calculations from the separation trials, the
head grades of the ore feed were determined to be 1003
ppm Sn, 174 ppm Ta, 107 ppm Nb, and 3222 ppm Rb.
Centrifugal Concentration Trials
Machine Set-Up and Settings
The centrifugal separator employed in the study is a lab-
oratory-sized L40 Falcon concentrator (Sepro Mineral
Systems, Canada). A 4” Semi-Batch (SB) bowl, facilitating
the injection of fluidization water in the retention zones of
the concentrate, was utilized. The two adjustable machine
parameters were the rotation speed and fluidization water
pressure, controlled by a speed controller and a flowrate
meter, respectively.
Dry ore was fed at a constant flow rate of 30 kg/h
through a vibratory feeder onto a sieve, representing 12%
of the maximum theoretical solid flow rate of the machine
(Sepro Mineral Systems Corp., 2018). Simultaneously,
water was sprayed onto the same sieve at a fixed flow rate of
70 kg/h. The resulting 30 wt.% solid pulp passed through
the sieve to reach the machine and enter the separator until
one kg of dry sample was utilized. At this juncture, flu-
idization water and rotation speed were halted to facilitate
concentrate collection. The collection involved merging
products from the inside of the bowl and the bottom of
the water jacket, which contained particles that successfully
passed through the fluidization water injection nozzles dur-
ing the separation. Each sample underwent a single concen-
tration step, constituting a roughing trial.
Chemical Analysis
The samples underwent pulverization using a planetary mill
to generate particles with sizes below 20μm. Subsequently,
the pulverized samples were analysed using a specialized
sample cup in conjunction with a Niton XL3t GOLDD+
Previous Page Next Page